Jennifer R. Brown, MD, PhD

Question 1. In your opinion, what were the most clinically relevant CLL data presented at ASCO this year?

Answer
Certainly the most important data on CLL presented this year were from the CLL14 study.¹ It was presented simultaneously with its publication in The New England Journal of Medicine² and the regimen had already been approved by the FDA [US Food and Drug Administration] a couple weeks earlier.³ This was a randomized trial comparing 1 year of venetoclax/obinutuzumab to 1 year of chlorambucil/obinutuzumab among CLL patients with comorbidities. The patient population was a fairly typical CLL first-line population in the sense that the median age was in the early 70s, and about 60% had IGHV [immunoglobulin heavy chain variable region]-unmutated status. Comorbidity score was above 6 in 80% to 90% of patients. These patients had more ‘other’ medical problems than were seen with any of the ibrutinib studies.

The regimen was pretty well tolerated. There was no tumor lysis attributable to the venetoclax. There was some neutropenia, but that was pretty well matched across treatment arms, as were the fatal adverse events during therapy.

The 2-year progression-free survival for venetoclax/obinutuzumab was 88%, which is quite excellent. Patients stopped the regimen as planned at 1 year and median follow-up after discontinuation of therapy was 17 months. There didn’t appear to be a significant drop off in PFS at the time of discontinuation.

Activity was very high in IGHV-unmutated patients, as well as [in] mutated patients. Patients with a 17p deletion did have a somewhat lower progression-free survival, which is typical even with novel agents, but they still had a very good progression-free survival. Undetectable minimal residual disease (MRD) in bone marrow was seen in 57% of patients in the venetoclax/obinutuzumab arm, which is also very good.

Question 2. Can you discuss the importance/significance of achieving a negative minimal residual disease (MRD) status after treatment for CLL?

Answer
MRD came to the fore initially as an end point with chemoimmunotherapy (CIT) after fludarabine, cyclophosphamide, and rituximab (FCR) or bendamustine and rituximab (BR) because the goal of those regimens is to achieve deep remission — complete remission ideally — and therapy is stopped after 6 months. In that setting, it was found that having undetectable MRD was strongly associated with progression-free survival as well as overall survival.

Now with venetoclax, which is given for timed durations in the frontline and early-relapse settings, the drug achieves deep remissions and we discontinue therapy, similar to what is done with CIT. We have also seen in the relapse setting from multiple venetoclax trials, that undetectable MRD strongly predicts progression-free survival. In the setting of venetoclax or CIT, achieving undetectable MRD is a strong predictor of outcome, benefit, and survival.

The part that becomes confusing is that with BTK [Bruton tyrosine kinase] and PI3 kinase inhibitors, the structure of the therapy is different. Patients go on the treatment with a plan for indefinite therapy, and few achieve complete remission. While partial-remission patients after venetoclax or CIT can also have undetectable MRD, usually, in that case, the partial remission is just based on small lymph nodes. In contrast, people who still have circulating disease, which is common with BTK inhibitors and PI3 kinase inhibitors, then measuring MRD doesn’t make sense.

Thus, with the BCR pathway inhibitors, you can have prolonged progression-free survival despite persistent disease, as long as you stay on the drug. In the setting of those drugs, or even when you add an antibody to those drugs, it is not so clear that achieving undetectable MRD is important. But with venetoclax, combination therapy that includes venetoclax or CIT — there, we know that achieving undetectable MRD strongly associates with outcomes.

Question 3. What is your view on the recent developments in the field of chimeric antigen receptor T-cell (CAR-T) therapies in CLL?

Answer
There is certainly a lot of renewed excitement about CAR-T cells in CLL. There had been excitement with early reports of durable remission among patients with a 17p deletion from the University of Pennsylvania, but follow-up reports showed response rates in CLL that were lower than in other diseases. Enthusiasm waned a bit.

In the last few years, there has been some renewed enthusiasm, with some studies combining ibrutinib with CAR-T cells, suggesting that perhaps ibrutinib is able to correct the underlying immune milieu and enhance collection and expansion of cells, and maybe reduce cytokine release syndrome (CRS). There has also been some enhanced toxicity with ibrutinib though, for example, cardiac deaths. It is going to be important to study in exactly what ways, and with what timing during therapy, ibrutinib or other BTK inhibitors enhance CAR-T cell therapy, versus increasing toxicity, in order to optimize use.

There are also recent data with JCAR17 without the use of ibrutinib.⁴ This is a somewhat different CD19 CAR-T product because it is produced at a defined ratio of CD4 and CD8 cells, which is not true of the others. Those data were presented at ASCO this year and looked encouraging. The rate of grade 3 or worse CRS was less than 10%, and grade 3 or higher neurotoxicity was about 20%. It looked like tolerability was better and response rates are higher as well. The overall response was about 82%, with about a 46% complete remission rate with a median follow-up of 9 months. That looks quite encouraging albeit, with short follow-up. The patient population studied was primarily exposed to ibrutinib and some had also [been given] venetoclax. Thus, people with aggressive disease seem to be responding well, but we need more data and trials, as numbers remain quite small.

Question 4. Real-world data seem to suggest that use of tyrosine kinase inhibitors (TKIs) for CLL is associated with high discontinuation rates because of adverse events. What are your strategies for monitoring and addressing adverse events in patients being treated with TKIs?

Answer
Communication is a key point. If you forewarn the patients that some of these things — like early diarrhea, or joint aches and pains — are likely to happen, they can be prepared and may be more tolerant. There are some strategies for GI side effects. For example, it may be better to take drugs at night rather than in the morning. It is also the case that many of the side effects tend to be worse in the first few months and will get better over time. I encourage patients to try to wait it out and give it a chance because oftentimes some of the side effects may go away.

Question 5. What impact does genome sequencing have on treatment selection, and how do you communicate with patients about genetic testing?

Answer
The most important genetic testing involves looking for a 17p deletion and/or TP53 mutation. It is clear that if those are found, patients need to get novel agent therapy and may benefit from combination clinical trials. There is some risk for earlier relapse. That testing is, therefore, generally indicated regardless of patient preference.

We do TP53 mutation testing in a panel that includes a number of other genes that are driver genes that are mutated in CLL. In that context, we learn what other mutations the patient may have in their tumor. There are some data that show the more mutations a patient with CLL has, the higher-risk the disease. In the era of CIT, each additional mutation is associated with increased risk of worse survival. That may not be as true in the era of novel agents, but at present, this remains unknown.

Outside the context of TP53, we don’t know a lot about other individual mutations. NOTCH1 is another gene that we commonly look for because it has been associated with an increased risk of Richter transformation. At the moment though, we have no strategies to prevent Richter transformation in NOTCH1-mutated patients. It is mostly a case of worrying rather than addressing.

Another aspect of sequencing is to look at IGHV mutational status. I view this as a very fundamental test because it really dictates the overall behavior of the disease course from diagnosis on. That is true regardless of whether patients are on therapy or not. Particularly when patients are not on therapy, you can see that the rate of progression is often strongly dependent on IGHV, and whether it is mutated (which is favorable), or unmutated (which is unfavorable). Furthermore, among young fit patients, a mutated IGHV is associated with potentially curable disease with FCR. These are patients where I still always recommend CIT, sometimes combined with ibrutinib based on a phase 2 study that we did here that results in undetectable MRD in bone marrow in 84% of patients.⁵ It is important to identify that these patients may have the opportunity for a very long-term durable remission — possibly [even] a cure.